AttributeError: 'AutoencoderKLOutput' object has no attribute 'mode'
#127
by cjohndesign - opened
https://colab.research.google.com/drive/1d-ib6cpmBzM1SIBEnImDA5GZAoeLOkFx?usp=sharing
On Scheduling and Visualisation
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
<ipython-input-46-5c3f5641a280> in <module>
1 im = Image.open('/content/drive/MyDrive/stablediff/lp/moth1.png').convert('RGB')
2 im = im.resize((512,512))
----> 3 encoded = pil_to_latent(im)
4 im
<ipython-input-45-e9e14c141363> in pil_to_latent(input_im)
6 with torch.no_grad():
7 latent = vae.encode(to_tensor_tfm(input_im).unsqueeze(0).to(torch_device)*2-1) # Note scaling
----> 8 return 0.18215 * latent.mode() # or .mean or .sample
9
10 def latents_to_pil(latents):
AttributeError: 'AutoencoderKLOutput' object has no attribute 'mode'```
Try
return 0.18215 * latent.latent_dist.mode()